No abstract

No abstract

No abstract

Convolutional neural network (CNN), a well-known machine learning algorithm, has been widely used in the field of computer vision for its amazing performance in image classification. With the rapid growth of applications based on CNN, various acceleration schemes have been proposed on FPGA, GPU and ASIC. In the implementation of these specific hardware accelerations, the most challenging part is the implementation of 2D convolution. To obtain a more efficient design of 2D convolution in CNN, this paper proposes a novel technique, singular value decomposition approximation (SVDA) to reduce the usage of resources. Experimental results show that the proposed SVDA hardware implementation can achieve a reduction in resources in the range of 14.46% to 37.8%, while the loss of classification accuracy is less than 1%.

Quality regulation is an important issue in platform governance. Quality regulation strategies can be targeted either to the platform or to the complementors. Different from a large number of previous studies on a single strategy, this paper investigates the combination of several quality regulation strategies for two-sided platforms. This paper constructs a profit maximization model to explore the optimal performance investment and price decisions of the platform under multiple quality regulation strategies. Based on the existing research, this study considers the proportion of high-quality complementors on the optimal performance decision, so that the two types of quality regulation strategies, platform-oriented and complementor-oriented, can be combined. This study finds that the optimal performance investment decision is influenced by the implementation effect of other quality regulation strategies for complementors. Therefore, platform owners should combine subsidies with quality regulation strategies when making performance investments. In addition, our results demonstrate that the combination of several quality regulation strategies can eliminate the problem caused by a certain single strategy, such as free-riding and closed platforms.